Detection and treatment of cancers using radiopharmaceuticals that selectively target cancers in human patients has been employed for several decades. Unfortunately, only a limited number of site-directed radiopharmaceuticals that exhibit highly specific in vivo localization in or near cancer cells are currently in routine use, as being approved by the United States Food and Drug Administration (FDA). There is a great deal of interest in developing new radioactive drugs due to the emergence of more sophisticated biomolecular carriers that have high affinity and high specificity for in vivo targeting of tumors. Several types of agents are being developed and have been investigated including monoclonal antibodies (MAbs), antibody fragments (F.sub.AB 's and (F.sub.AB).sub.2 's), receptor-avid peptides [Bushbaum, 1995; Fischman et al., 1993; Schubiger et al. 1996].
The potential utility of using radiolabeled receptor-avid peptides for producing radiopharmaceuticals is best exemplified by .sup.111 In-DTPA-conjugated octreotide (an FDA approved diagnostic imaging agent, Octreoscan.RTM., marketed in the United States. by Mallinckrodt Medical, Inc.) [Lowbertz et al. 1994]. This radiopharmaceutical is an .sup.111 In-DTPA conjugate of Octreotide, a small peptide analogue of the human hormone somatostatin. This drug specifically binds to somatostatin receptors that are over-expressed on neuroendocrine cancers (e.g., carcinoid Ca, neuroblastoma, etc.) as well as others [Krenning et al., 1994]. Since indium-111 (.sup.111 In) is not the ideal radionuclide for scintigraphic imaging, other somatostatin analogues and other receptor-avid biomolecules that are labeled with .sup.99m Tc (the optimal radionuclide for diagnostic imaging) are being studied and developed [Eckelman, 1995; Vallabhajosula et al., 1996].
Bombesin (BBN) is a 14 amino acid peptide that is an analogue of human gastrin releasing peptide (GRP) that binds to GRP receptors with high specificity and has an affinity similar to GRP [Davis et al., 1992]. GRP receptors have been shown to be over-expressed or uniquely expressed on several types of cancer cells. Binding of GRP receptor agonists (also autocrine factors) increases the rate of cell division of these cancer cells. For this reason, a great deal of work has been, and is being pursued to develop BBN or GRP analogues that are antagonists [Davis et al., 1992; Hoffken, 1994; Moody et al., 1996; Coy et al., 1988; Cai et al., 1994]. These antagonists are designed to competitively inhibit endogenous GRP binding to GRP receptors and reduce the rate of cancer cell proliferation [Hoffken, 1994]. Treatment of cancers using these antagonists with these non-radioactive peptides requires chronic injection regimens (e.g., daily, using large quantities of the drug).
In designing an effective receptor-avid radiopharmaceutical for use as a diagnostic or therapeutic agent for cancer, it is important that the drug have appropriate in vivo targeting and pharmacokinetic properties [Fritzberg et al., 1992; Eckelman et al., 1993]. For example, it is essential that the radiolabeled receptor-avid peptide have high specific uptake by the cancer cells (e.g., via GRP receptors). In addition, it is necessary that once the radionuclide localizes at a tumor site, it must remain there for an extended time to deliver a highly localized radiation dose to the tumor. In order to achieve sufficiently high specific uptake of radiolabeled BBN analogues in tumors, the binding affinity of promising derivatives must be high (i.e., K.sub.d.congruent.1-5 nmolar or less) with prolonged retention of radioactivity [Eckelman et al., 1995; Eckelman, et al., 1993]. Work with .sup.125 1-BBN derivatives has shown, however, that for cancer cells that bind the .sup.125 I-BBN derivatives (whether they be agonists or antagonists), the radioactivity is either washed off or expelled from the cells (in vitro) at a rapid rate [Hoffman et al., 1997]. Thus, these types of derivatives have a low probability of remaining "trapped" at the tumor site (in vivo) sufficiently long to be effective therapeutic or diagnostic agents.
Developing radiolabeled peptides that are cleared efficiently from normal tissues is also an important and especially critical factor for therapeutic agents. When labeling biomolecules (e.g., MAb, F.sub.AB 's or peptides) with metallic radionuclides (via a chelate conjugation), a large percentage of the metallic radionuclide (in some chemical form) usually becomes "trapped" in either the kidney or liver parenchyma (i.e., is not excreted into the urine or bile) [Duncan et al., 1997; Mattes, 1995]. For the smaller radiolabeled biomolecules (i.e., peptides or F.sub.AB 's), the major route of clearance of activity is through the kidneys which in turn retain high levels of the radioactive metal (i.e., normally &gt;10-15% of the injected dose) [Duncan et al., 1997]. This presents a major problem that must be overcome in the development of therapeutic agents that incorporate metallic radionuclides, otherwise the radiation dose to the kidneys would be excessive. For example, .sup.111 In-octreotide, the FDA approved diagnostic agent, exhibits high uptake and retention in kidneys of patients [Eckelman et al., 1995]. Even though the radiation dose to the kidneys is higher than desirable, it is tolerable in that it is a diagnostic radiopharmaceutical (it does not emit alpha- or beta-particles), and the renal dose does not produce observable radiation induced damage to the organ.
It has now been found that conjugating BBN derivatives which are agonists in non-metallated conjugates whichthat exhibit binding affinities to GRP receptors that are either similar to or approximately an order of magnitude lower than the parent BBN derivative. [Li et al., 1996a] These data coupled with our recent results show that it is now possible to add radiometal chelates to BBN analogues, which are agonists, and retain GRP receptor binding affinities that are sufficiently high (i.e., approx. 1-5 nmolar K.sub.d 's) for further development as potential radiopharmaceuticals. These agonist conjugates are transported intracellularly after binding to cell surface GRP receptors and retained inside of the cells for extended time periods. In addition, in vivo studies in normal mice have shown that retention of the radioactive metal in the kidneys was low (i.e., &lt;5%) with the majority of the radioactivity excreted into the urine.
According to one aspect of the present invention, there is provided a BBN conjugate consisting of essentially a radio-metal chelate covalently appended to the receptor binding region of BBN [e.g., BBN(8-14)] to form radiolabeled BBN analogues that have high specific binding affinities with GRP receptors. These analogues are retained for long times inside of GRP expressing cancer cells. Furthermore, their clearance from the bloodstream, into the urine with minimal kidney retention, is efficient. Preferably, the radiometals are selected from .sup.99m Tc, .sup.186/188 Re, .sup.105 Rh, .sup.153 Sm, .sup.166 Ho, .sup.90 Y or .sup.199 Au, all of which hold the potential for diagnostic (i.e., .sup.99m Tc) or therapeutic (i.e., .sup.186/188 Re, .sup.105 Rh, .sup.153 Sm, .sup.166 Ho, .sup.90 Y, and .sup.199 Au) utility in cancer patients [Schubiger et al, 1996; Eckelman, 1995; Troutner, 1978].